Senior Design

twelve <- read.csv("C:/Users/farlc_000/Documents/2012.csv")
fourteen <- read.csv("C:/Users/farlc_000/Documents/2014.csv")
sixteen <- read.csv("C:/Users/farlc_000/Documents/2016.csv")

library(randomForest)
library(caret)

fit16 = randomForest(data = sixteen[,-(1:4)], DemocraticTwoPartyVoteshare ~.)

varImpPlot(fit16, type=2)
VI <- importance(fit16)
write.csv(VI, file = "VI16.csv", row.names = T)

sorted <- read.csv("VI16.csv")
barplot(sorted$IncNodePurity[1:15], horiz = T, names.arg = sorted$Var[1:15])

fit12 = randomForest(data = twelve[,-(1:4)], DemocraticTwoPartyVoteshare ~.)

varImpPlot(fit12, type=2)

VI <- importance(fit12)
write.csv(VI, file = "VI12.csv", row.names = T)

fit14 = randomForest(data = fourteen[,-(1:4)], DemocraticTwoPartyVoteshare ~.)

varImpPlot(fit14, type=2)

VI <- importance(fit14)
write.csv(VI, file = "VI14.csv", row.names = T)

data1416 <- rbind(fourteen, sixteen)
AllData <- rbind(twelve, data1416)

X <- split(AllData, AllData$X)
cali <- X[[1]]
penn <- X[[2]]
tx <- X[[3]]
wash <- X[[4]]

fitcali <- randomForest(data = cali[,-(1:4)], DemocraticTwoPartyVoteshare ~.)

varImpPlot(fitcali, type = 2)

fitwash <- randomForest(data = wash[,-(1:4)], DemocraticTwoPartyVoteshare ~.)

varImpPlot(fitwash, type = 2)

fittx <- randomForest(data = tx[,-(1:4)], DemocraticTwoPartyVoteshare ~.)

varImpPlot(fittx, type = 2)

fitpenn <- randomForest(data = penn[,-(1:4)], DemocraticTwoPartyVoteshare ~.)

varImpPlot(fitpenn, type = 2)

VI <- importance(fitpenn)
write.csv(VI, file = "VIPA.csv", row.names = T)
VI <- importance(fittx)
write.csv(VI, file = "VITX.csv", row.names = T)
VI <- importance(fitwash)
write.csv(VI, file = "VIWA.csv", row.names = T)
VI <- importance(fitcali)
write.csv(VI, file = "VICA.csv", row.names = T)

WAneighbor <- read.csv("~/Senior Year/Senior Design/Washington Neighbors.csv", header = T, row.names = 1)
PAneighbor <- read.csv("~/Senior Year/Senior Design/Pennsylvania Neighbors.csv", header = T, row.names = 1)
TXneighbor <- read.csv("~/Senior Year/Senior Design/Texas Neighbors v2.csv", header = T, row.names = 1)
CAneighbor <- read.csv("~/Senior Year/Senior Design/California Neighbors.csv", header = T, row.names = 1)

cali$DistrictNum <- cali$GEO.id2 - 600
wash$DistrictNum <- wash$GEO.id2 - 5300
penn$DistrictNum <- penn$GEO.id2 - 4200
tx$DistrictNum <- tx$GEO.id2 - 4800
cali2012 <- cali[(1:53),]
wash2012 <- wash[(1:10),]
penn2012 <- penn[(1:18),]
tx2012 <- tx[(1:36),]

neighbordifference <- function(district, demo, neighbor){ #assumes both are vectors of equal size
  n <- sum(neighbor)
  nbrsdem <- demo[which(neighbor)]
  differences <- abs(demo[district] - nbrsdem)
  neighbordifference <- sum(differences) / n
}
num_districts <- length(cali2012$GEO.id)
agdifferencesCA <- rep(NA, num_districts)
vetdifferencesCA <- rep(NA, num_districts)
blackdifferencesCA <- rep(NA, num_districts)
incomedifferencesCA <- rep(NA, num_districts)
foreigndifferencesCA <- rep(NA, num_districts)
unempdifferencesCA <- rep(NA, num_districts)
youngdifferencesCA <- rep(NA, num_districts)
somecollegedifferencesCA <- rep(NA, num_districts)
whitedifferencesCA <- rep(NA, num_districts)
tworacedifferencesCA <- rep(NA, num_districts)
elderlydifferencesCA <- rep(NA, num_districts)
tradedifferencesCA <- rep(NA, num_districts)
englishdifferencesCA <- rep(NA, num_districts)
marrieddifferencesCA <- rep(NA, num_districts)
hispanicdifferencesCA <- rep(NA, num_districts)
enrolleddifferencesCA <- rep(NA, num_districts)
for (i in 1:num_districts){
  agdifferencesCA[i] <- neighbordifference(i, cali2012$HC03_EC50, unname(CAneighbor[,i]))
  vetdifferencesCA[i] <- neighbordifference(i, cali2012$HC03_SC101, unname(CAneighbor[,i]))
  blackdifferencesCA[i] <- neighbordifference(i, cali2012$HC03_DH95, unname(CAneighbor[,i]))
  incomedifferencesCA[i] <- neighbordifference(i, cali2012$HC01_EC118, unname(CAneighbor[,i]))
  foreigndifferencesCA[i] <- neighbordifference(i, cali2012$HC03_SC140, unname(CAneighbor[,i]))
  unempdifferencesCA[i] <- neighbordifference(i, cali2012$HC03_EC12, unname(CAneighbor[,i]))
  youngdifferencesCA[i] <- neighbordifference(i, cali2012$HC03_EST_ASB02, unname(CAneighbor[,i]))
  somecollegedifferencesCA[i] <- neighbordifference(i, cali2012$HC03_SC89, unname(CAneighbor[,i]))
  whitedifferencesCA[i] <- neighbordifference(i, cali2012$HC03_DH94, unname(CAneighbor[,i]))
  tworacedifferencesCA[i] <- neighbordifference(i, cali2012$HC03_SC101, unname(CAneighbor[,i]))
  elderlydifferencesCA[i] <- neighbordifference(i, cali2012$HC03_DH102, unname(CAneighbor[,i]))
  tradedifferencesCA[i] <- neighbordifference(i, cali2012$HC03_EC53, unname(CAneighbor[,i]))
  englishdifferencesCA[i] <- neighbordifference(i, cali2012$HC03_SC171, unname(CAneighbor[,i]))
  marrieddifferencesCA[i] <- neighbordifference(i, cali2012$HC03_SC45, unname(CAneighbor[,i]))
  hispanicdifferencesCA[i] <- neighbordifference(i, cali2012$HC03_DH88, unname(CAneighbor[,i]))
  enrolleddifferencesCA[i] <- neighbordifference(i, cali2012$HC03_SC81, unname(CAneighbor[,i]))
}
num_districts <- length(wash2012$DistrictNum)
agdifferencesWA <- rep(NA, num_districts)
vetdifferencesWA <- rep(NA, num_districts)
blackdifferencesWA <- rep(NA, num_districts)
incomedifferencesWA <- rep(NA, num_districts)
foreigndifferencesWA <- rep(NA, num_districts)
unempdifferencesWA <- rep(NA, num_districts)
youngdifferencesWA <- rep(NA, num_districts)
somecollegedifferencesWA <- rep(NA, num_districts)
whitedifferencesWA <- rep(NA, num_districts)
tworacedifferencesWA <- rep(NA, num_districts)
elderlydifferencesWA <- rep(NA, num_districts)
tradedifferencesWA <- rep(NA, num_districts)
englishdifferencesWA <- rep(NA, num_districts)
marrieddifferencesWA <- rep(NA, num_districts)
hispanicdifferencesWA <- rep(NA, num_districts)
enrolleddifferencesWA <- rep(NA, num_districts)
for (i in 1:num_districts){
  agdifferencesWA[i] <- neighbordifference(i, wash2012$HC03_EC50, unname(WAneighbor[,i]))
  vetdifferencesWA[i] <- neighbordifference(i, wash2012$HC03_SC101, unname(WAneighbor[,i]))
  blackdifferencesWA[i] <- neighbordifference(i, wash2012$HC03_DH95, unname(WAneighbor[,i]))
  incomedifferencesWA[i] <- neighbordifference(i, wash2012$HC01_EC118, unname(WAneighbor[,i]))
  foreigndifferencesWA[i] <- neighbordifference(i, wash2012$HC03_SC140, unname(WAneighbor[,i]))
  unempdifferencesWA[i] <- neighbordifference(i, wash2012$HC03_EC12, unname(WAneighbor[,i]))
  youngdifferencesWA[i] <- neighbordifference(i, wash2012$HC03_EST_ASB02, unname(WAneighbor[,i]))
  somecollegedifferencesWA[i] <- neighbordifference(i, wash2012$HC03_SC89, unname(WAneighbor[,i]))
  whitedifferencesWA[i] <- neighbordifference(i, wash2012$HC03_DH94, unname(WAneighbor[,i]))
  tworacedifferencesWA[i] <- neighbordifference(i, wash2012$HC03_SC101, unname(WAneighbor[,i]))
  elderlydifferencesWA[i] <- neighbordifference(i, wash2012$HC03_DH102, unname(WAneighbor[,i]))
  tradedifferencesWA[i] <- neighbordifference(i, wash2012$HC03_EC53, unname(WAneighbor[,i]))
  englishdifferencesWA[i] <- neighbordifference(i, wash2012$HC03_SC171, unname(WAneighbor[,i]))
  marrieddifferencesWA[i] <- neighbordifference(i, wash2012$HC03_SC45, unname(WAneighbor[,i]))
  hispanicdifferencesWA[i] <- neighbordifference(i, wash2012$HC03_DH88, unname(WAneighbor[,i]))
  enrolleddifferencesWA[i] <- neighbordifference(i, wash2012$HC03_SC81, unname(WAneighbor[,i]))
}
num_districts <- length(penn2012$DistrictNum)
agdifferencesPA <- rep(NA, num_districts)
vetdifferencesPA <- rep(NA, num_districts)
blackdifferencesPA <- rep(NA, num_districts)
incomedifferencesPA <- rep(NA, num_districts)
foreigndifferencesPA <- rep(NA, num_districts)
unempdifferencesPA <- rep(NA, num_districts)
youngdifferencesPA <- rep(NA, num_districts)
somecollegedifferencesPA <- rep(NA, num_districts)
whitedifferencesPA <- rep(NA, num_districts)
tworacedifferencesPA <- rep(NA, num_districts)
elderlydifferencesPA <- rep(NA, num_districts)
tradedifferencesPA <- rep(NA, num_districts)
englishdifferencesPA <- rep(NA, num_districts)
marrieddifferencesPA <- rep(NA, num_districts)
hispanicdifferencesPA <- rep(NA, num_districts)
enrolleddifferencesPA <- rep(NA, num_districts)
for (i in 1:num_districts){
  agdifferencesPA[i] <- neighbordifference(i, penn2012$HC03_EC50, unname(PAneighbor[,i]))
  vetdifferencesPA[i] <- neighbordifference(i, penn2012$HC03_SC101, unname(PAneighbor[,i]))
  blackdifferencesPA[i] <- neighbordifference(i, penn2012$HC03_DH95, unname(PAneighbor[,i]))
  incomedifferencesPA[i] <- neighbordifference(i, penn2012$HC01_EC118, unname(PAneighbor[,i]))
  foreigndifferencesPA[i] <- neighbordifference(i, penn2012$HC03_SC140, unname(PAneighbor[,i]))
  unempdifferencesPA[i] <- neighbordifference(i, penn2012$HC03_EC12, unname(PAneighbor[,i]))
  youngdifferencesPA[i] <- neighbordifference(i, penn2012$HC03_EST_ASB02, unname(PAneighbor[,i]))
  somecollegedifferencesPA[i] <- neighbordifference(i, penn2012$HC03_SC89, unname(PAneighbor[,i]))
  whitedifferencesPA[i] <- neighbordifference(i, penn2012$HC03_DH94, unname(PAneighbor[,i]))
  tworacedifferencesPA[i] <- neighbordifference(i, penn2012$HC03_SC101, unname(PAneighbor[,i]))
  elderlydifferencesPA[i] <- neighbordifference(i, penn2012$HC03_DH102, unname(PAneighbor[,i]))
  tradedifferencesPA[i] <- neighbordifference(i, penn2012$HC03_EC53, unname(PAneighbor[,i]))
  englishdifferencesPA[i] <- neighbordifference(i, penn2012$HC03_SC171, unname(PAneighbor[,i]))
  marrieddifferencesPA[i] <- neighbordifference(i, penn2012$HC03_SC45, unname(PAneighbor[,i]))
  hispanicdifferencesPA[i] <- neighbordifference(i, penn2012$HC03_DH88, unname(PAneighbor[,i]))
  enrolleddifferencesPA[i] <- neighbordifference(i, penn2012$HC03_SC81, unname(PAneighbor[,i]))
}
num_districts <- length(tx2012$DistrictNum)
agdifferencesTX <- rep(NA, num_districts)
vetdifferencesTX <- rep(NA, num_districts)
blackdifferencesTX <- rep(NA, num_districts)
incomedifferencesTX <- rep(NA, num_districts)
foreigndifferencesTX <- rep(NA, num_districts)
unempdifferencesTX <- rep(NA, num_districts)
youngdifferencesTX <- rep(NA, num_districts)
somecollegedifferencesTX <- rep(NA, num_districts)
whitedifferencesTX <- rep(NA, num_districts)
tworacedifferencesTX <- rep(NA, num_districts)
elderlydifferencesTX <- rep(NA, num_districts)
tradedifferencesTX <- rep(NA, num_districts)
englishdifferencesTX <- rep(NA, num_districts)
marrieddifferencesTX <- rep(NA, num_districts)
hispanicdifferencesTX <- rep(NA, num_districts)
enrolleddifferencesTX <- rep(NA, num_districts)
for (i in 1:num_districts){
  agdifferencesTX[i] <- neighbordifference(i, tx2012$HC03_EC50, unname(TXneighbor[,i]))
  vetdifferencesTX[i] <- neighbordifference(i, tx2012$HC03_SC101, unname(TXneighbor[,i]))
  blackdifferencesTX[i] <- neighbordifference(i, tx2012$HC03_DH95, unname(TXneighbor[,i]))
  incomedifferencesTX[i] <- neighbordifference(i, tx2012$HC01_EC118, unname(TXneighbor[,i]))
  foreigndifferencesTX[i] <- neighbordifference(i, tx2012$HC03_SC140, unname(TXneighbor[,i]))
  unempdifferencesTX[i] <- neighbordifference(i, tx2012$HC03_EC12, unname(TXneighbor[,i]))
  youngdifferencesTX[i] <- neighbordifference(i, tx2012$HC03_EST_ASB02, unname(TXneighbor[,i]))
  somecollegedifferencesTX[i] <- neighbordifference(i, tx2012$HC03_SC89, unname(TXneighbor[,i]))
  whitedifferencesTX[i] <- neighbordifference(i, tx2012$HC03_DH94, unname(TXneighbor[,i]))
  tworacedifferencesTX[i] <- neighbordifference(i, tx2012$HC03_SC101, unname(TXneighbor[,i]))
  elderlydifferencesTX[i] <- neighbordifference(i, tx2012$HC03_DH102, unname(TXneighbor[,i]))
  tradedifferencesTX[i] <- neighbordifference(i, tx2012$HC03_EC53, unname(TXneighbor[,i]))
  englishdifferencesTX[i] <- neighbordifference(i, tx2012$HC03_SC171, unname(TXneighbor[,i]))
  marrieddifferencesTX[i] <- neighbordifference(i, tx2012$HC03_SC45, unname(TXneighbor[,i]))
  hispanicdifferencesTX[i] <- neighbordifference(i, tx2012$HC03_DH88, unname(TXneighbor[,i]))
  enrolleddifferencesTX[i] <- neighbordifference(i, tx2012$HC03_SC81, unname(TXneighbor[,i]))
}
agconfCA <- t.test(agdifferencesCA, conf.level = 0.95)$conf.int
agconfWA <- t.test(agdifferencesWA, conf.level = 0.95)$conf.int
agconfPA <- t.test(agdifferencesPA, conf.level = 0.95)$conf.int
agconfTX <- t.test(agdifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(agdifferencesCA), mean(agdifferencesWA), mean(agdifferencesPA), mean(agdifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Percentage of Agriculture Workers", 4)
lowerconf <- c(agconfCA[1], agconfWA[1], agconfPA[1], agconfTX[1])
upperconf <- c(agconfCA[2], agconfWA[2], agconfPA[2], agconfTX[2])
agdf <- data.frame(states, means, lowerconf, upperconf, label)
vetconfCA <- t.test(vetdifferencesCA, conf.level = 0.95)$conf.int
vetconfWA <- t.test(vetdifferencesWA, conf.level = 0.95)$conf.int
vetconfPA <- t.test(vetdifferencesPA, conf.level = 0.95)$conf.int
vetconfTX <- t.test(vetdifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(vetdifferencesCA), mean(vetdifferencesWA), mean(vetdifferencesPA), mean(vetdifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Percentage of Veterans", 4)
lowerconf <- c(vetconfCA[1], vetconfWA[1], vetconfPA[1], vetconfTX[1])
upperconf <- c(vetconfCA[2], vetconfWA[2], vetconfPA[2], vetconfTX[2])
vetdf <- data.frame(states, means, lowerconf, upperconf, label)
blackconfCA <- t.test(blackdifferencesCA, conf.level = 0.95)$conf.int
blackconfWA <- t.test(blackdifferencesWA, conf.level = 0.95)$conf.int
blackconfPA <- t.test(blackdifferencesPA, conf.level = 0.95)$conf.int
blackconfTX <- t.test(blackdifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(blackdifferencesCA), mean(blackdifferencesWA), mean(blackdifferencesPA), mean(blackdifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage Population Black", 4)
lowerconf <- c(blackconfCA[1], blackconfWA[1], blackconfPA[1], blackconfTX[1])
upperconf <- c(blackconfCA[2], blackconfWA[2], blackconfPA[2], blackconfTX[2])
blackdf <- data.frame(states, means, lowerconf, upperconf, label)
incomeconfCA <- t.test(incomedifferencesCA, conf.level = 0.95)$conf.int
incomeconfWA <- t.test(incomedifferencesWA, conf.level = 0.95)$conf.int
incomeconfPA <- t.test(incomedifferencesPA, conf.level = 0.95)$conf.int
incomeconfTX <- t.test(incomedifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(incomedifferencesCA), mean(incomedifferencesWA), mean(incomedifferencesPA), mean(incomedifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Per Capita Income Difference", 4)
lowerconf <- c(incomeconfCA[1], incomeconfWA[1], incomeconfPA[1], incomeconfTX[1])
upperconf <- c(incomeconfCA[2], incomeconfWA[2], incomeconfPA[2], incomeconfTX[2])
incomedf <- data.frame(states, means, lowerconf, upperconf, label)
foreignconfCA <- t.test(foreigndifferencesCA, conf.level = 0.95)$conf.int
foreignconfWA <- t.test(foreigndifferencesWA, conf.level = 0.95)$conf.int
foreignconfPA <- t.test(foreigndifferencesPA, conf.level = 0.95)$conf.int
foreignconfTX <- t.test(foreigndifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(foreigndifferencesCA), mean(foreigndifferencesWA), mean(foreigndifferencesPA), mean(foreigndifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage Foreign Born", 4)
lowerconf <- c(foreignconfCA[1], foreignconfWA[1], foreignconfPA[1], foreignconfTX[1])
upperconf <- c(foreignconfCA[2], foreignconfWA[2], foreignconfPA[2], foreignconfTX[2])
foreigndf <- data.frame(states, means, lowerconf, upperconf, label)
unempconfCA <- t.test(unempdifferencesCA, conf.level = 0.95)$conf.int
unempconfWA <- t.test(unempdifferencesWA, conf.level = 0.95)$conf.int
unempconfPA <- t.test(unempdifferencesPA, conf.level = 0.95)$conf.int
unempconfTX <- t.test(unempdifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(unempdifferencesCA), mean(unempdifferencesWA), mean(unempdifferencesPA), mean(unempdifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Unemployment Rate", 4)
lowerconf <- c(unempconfCA[1], unempconfWA[1], unempconfPA[1], unempconfTX[1])
upperconf <- c(unempconfCA[2], unempconfWA[2], unempconfPA[2], unempconfTX[2])
unempdf <- data.frame(states, means, lowerconf, upperconf, label)
youngconfCA <- t.test(youngdifferencesCA, conf.level = 0.95)$conf.int
youngconfWA <- t.test(youngdifferencesWA, conf.level = 0.95)$conf.int
youngconfPA <- t.test(youngdifferencesPA, conf.level = 0.95)$conf.int
youngconfTX <- t.test(youngdifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(youngdifferencesCA), mean(youngdifferencesWA), mean(youngdifferencesPA), mean(youngdifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage Aged 20-34", 4)
lowerconf <- c(youngconfCA[1], youngconfWA[1], youngconfPA[1], youngconfTX[1])
upperconf <- c(youngconfCA[2], youngconfWA[2], youngconfPA[2], youngconfTX[2])
youngdf <- data.frame(states, means, lowerconf, upperconf, label)
somecollegeconfCA <- t.test(somecollegedifferencesCA, conf.level = 0.95)$conf.int
somecollegeconfWA <- t.test(somecollegedifferencesWA, conf.level = 0.95)$conf.int
somecollegeconfPA <- t.test(somecollegedifferencesPA, conf.level = 0.95)$conf.int
somecollegeconfTX <- t.test(somecollegedifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(somecollegedifferencesCA), mean(somecollegedifferencesWA), mean(somecollegedifferencesPA), mean(somecollegedifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage with Some College but No Degree", 4)
lowerconf <- c(somecollegeconfCA[1], somecollegeconfWA[1], somecollegeconfPA[1], somecollegeconfTX[1])
upperconf <- c(somecollegeconfCA[2], somecollegeconfWA[2], somecollegeconfPA[2], somecollegeconfTX[2])
somecollegedf <- data.frame(states, means, lowerconf, upperconf, label)
whiteconfCA <- t.test(whitedifferencesCA, conf.level = 0.95)$conf.int
whiteconfWA <- t.test(whitedifferencesWA, conf.level = 0.95)$conf.int
whiteconfPA <- t.test(whitedifferencesPA, conf.level = 0.95)$conf.int
whiteconfTX <- t.test(whitedifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(whitedifferencesCA), mean(whitedifferencesWA), mean(whitedifferencesPA), mean(whitedifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage White", 4)
lowerconf <- c(whiteconfCA[1], whiteconfWA[1], whiteconfPA[1], whiteconfTX[1])
upperconf <- c(whiteconfCA[2], whiteconfWA[2], whiteconfPA[2], whiteconfTX[2])
whitedf <- data.frame(states, means, lowerconf, upperconf, label)
tworaceconfCA <- t.test(tworacedifferencesCA, conf.level = 0.95)$conf.int
tworaceconfWA <- t.test(tworacedifferencesWA, conf.level = 0.95)$conf.int
tworaceconfPA <- t.test(tworacedifferencesPA, conf.level = 0.95)$conf.int
tworaceconfTX <- t.test(tworacedifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(tworacedifferencesCA), mean(tworacedifferencesWA), mean(tworacedifferencesPA), mean(tworacedifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage Biracial", 4)
lowerconf <- c(tworaceconfCA[1], tworaceconfWA[1], tworaceconfPA[1], tworaceconfTX[1])
upperconf <- c(tworaceconfCA[2], tworaceconfWA[2], tworaceconfPA[2], tworaceconfTX[2])
tworacedf <- data.frame(states, means, lowerconf, upperconf, label)
elderlyconfCA <- t.test(elderlydifferencesCA, conf.level = 0.95)$conf.int
elderlyconfWA <- t.test(elderlydifferencesWA, conf.level = 0.95)$conf.int
elderlyconfPA <- t.test(elderlydifferencesPA, conf.level = 0.95)$conf.int
elderlyconfTX <- t.test(elderlydifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(elderlydifferencesCA), mean(elderlydifferencesWA), mean(elderlydifferencesPA), mean(elderlydifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage of Elderly Under the Poverty Line", 4)
lowerconf <- c(elderlyconfCA[1], elderlyconfWA[1], elderlyconfPA[1], elderlyconfTX[1])
upperconf <- c(elderlyconfCA[2], elderlyconfWA[2], elderlyconfPA[2], elderlyconfTX[2])
elderlydf <- data.frame(states, means, lowerconf, upperconf, label)
tradeconfCA <- t.test(tradedifferencesCA, conf.level = 0.95)$conf.int
tradeconfWA <- t.test(tradedifferencesWA, conf.level = 0.95)$conf.int
tradeconfPA <- t.test(tradedifferencesPA, conf.level = 0.95)$conf.int
tradeconfTX <- t.test(tradedifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(tradedifferencesCA), mean(tradedifferencesWA), mean(tradedifferencesPA), mean(tradedifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage Employed in Wholesale Trade", 4)
lowerconf <- c(tradeconfCA[1], tradeconfWA[1], tradeconfPA[1], tradeconfTX[1])
upperconf <- c(tradeconfCA[2], tradeconfWA[2], tradeconfPA[2], tradeconfTX[2])
tradedf <- data.frame(states, means, lowerconf, upperconf, label)
englishconfCA <- t.test(englishdifferencesCA, conf.level = 0.95)$conf.int
englishconfWA <- t.test(englishdifferencesWA, conf.level = 0.95)$conf.int
englishconfPA <- t.test(englishdifferencesPA, conf.level = 0.95)$conf.int
englishconfTX <- t.test(englishdifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(englishdifferencesCA), mean(englishdifferencesWA), mean(englishdifferencesPA), mean(englishdifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Percentage Speaking English at Home Difference", 4)
lowerconf <- c(englishconfCA[1], englishconfWA[1], englishconfPA[1], englishconfTX[1])
upperconf <- c(englishconfCA[2], englishconfWA[2], englishconfPA[2], englishconfTX[2])
englishdf <- data.frame(states, means, lowerconf, upperconf, label)
marriedconfCA <- t.test(marrieddifferencesCA, conf.level = 0.95)$conf.int
marriedconfWA <- t.test(marrieddifferencesWA, conf.level = 0.95)$conf.int
marriedconfPA <- t.test(marrieddifferencesPA, conf.level = 0.95)$conf.int
marriedconfTX <- t.test(marrieddifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(marrieddifferencesCA), mean(marrieddifferencesWA), mean(marrieddifferencesPA), mean(marrieddifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Percentage of Married Women Difference", 4)
lowerconf <- c(marriedconfCA[1], marriedconfWA[1], marriedconfPA[1], marriedconfTX[1])
upperconf <- c(marriedconfCA[2], marriedconfWA[2], marriedconfPA[2], marriedconfTX[2])
marrieddf <- data.frame(states, means, lowerconf, upperconf, label)
hispanicconfCA <- t.test(hispanicdifferencesCA, conf.level = 0.95)$conf.int
hispanicconfWA <- t.test(hispanicdifferencesWA, conf.level = 0.95)$conf.int
hispanicconfPA <- t.test(hispanicdifferencesPA, conf.level = 0.95)$conf.int
hispanicconfTX <- t.test(hispanicdifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(hispanicdifferencesCA), mean(hispanicdifferencesWA), mean(hispanicdifferencesPA), mean(hispanicdifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Difference in Percentage Population Hispanic", 4)
lowerconf <- c(hispanicconfCA[1], hispanicconfWA[1], hispanicconfPA[1], hispanicconfTX[1])
upperconf <- c(hispanicconfCA[2], hispanicconfWA[2], hispanicconfPA[2], hispanicconfTX[2])
hispanicdf <- data.frame(states, means, lowerconf, upperconf, label)
enrolledconfCA <- t.test(enrolleddifferencesCA, conf.level = 0.95)$conf.int
enrolledconfWA <- t.test(enrolleddifferencesWA, conf.level = 0.95)$conf.int
enrolledconfPA <- t.test(enrolleddifferencesPA, conf.level = 0.95)$conf.int
enrolledconfTX <- t.test(enrolleddifferencesTX, conf.level = 0.95)$conf.int
means <- c(mean(enrolleddifferencesCA), mean(enrolleddifferencesWA), mean(enrolleddifferencesPA), mean(enrolleddifferencesTX))
states <- c("California", "Washington", "Pennsylvania", "Texas")
label <- rep("Percent Enrolled in College Difference", 4)
lowerconf <- c(enrolledconfCA[1], enrolledconfWA[1], enrolledconfPA[1], enrolledconfTX[1])
upperconf <- c(enrolledconfCA[2], enrolledconfWA[2], enrolledconfPA[2], enrolledconfTX[2])
enrolleddf <- data.frame(states, means, lowerconf, upperconf, label)

ggplot(vetdf, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1))

vetag <- rbind(agdf, vetdf)
ggplot(vetag, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1))

ggplot(agdf, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1))

ggplot(blackdf, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1))

ggplot(incomedf, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1))

plot1 <- do.call("rbind", list(blackdf, agdf, vetdf))
plot2 <- do.call("rbind", list(unempdf, youngdf, somecollegedf, whitedf))
plot3 <- do.call("rbind", list(tworacedf, elderlydf, tradedf, englishdf))
plot4 <- do.call("rbind", list(marrieddf, hispanicdf, enrolleddf))

ggplot(plot1, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1)) +
  geom_hline(yintercept = 0)

ggplot(plot2, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1)) +
  geom_hline(yintercept = 0)

ggplot(plot3, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1)) +
  geom_hline(yintercept = 0)

ggplot(plot4, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1)) +
  geom_hline(yintercept = 0)

englishAllDistricts <- c(englishdifferencesCA, englishdifferencesPA, englishdifferencesTX, englishdifferencesWA)
allStates <- c(rep("California",53), rep("Pennsylvania",18), rep("Texas",36), rep("Washington", 10))
district <- c(cali2012$DistrictNum, penn2012$DistrictNum, tx2012$DistrictNum, wash2012$DistrictNum)
allCommission <- c(rep("Independent", 63), rep("Partisan", 54))
englishBoxDF <- data.frame(englishAllDistricts, allStates, district)
check_outlier <- function(v, coef=1.51){
  quantiles <- quantile(v,probs=c(0.25,0.75))
  IQR <- quantiles[2]-quantiles[1]
  res <- v < (quantiles[1]-coef*IQR)|v > (quantiles[2]+coef*IQR)
  return(res)
}

library(dplyr)
commissionEnglish <- englishBoxDF %>%
  mutate(commission = ifelse(allStates == "California", "Independent", ifelse(allStates == "Washington", "Independent", "Partisan")))
independentEnglish <- commissionEnglish %>%
  filter(commission == "Independent") %>%
  mutate(outliers = ifelse(check_outlier(englishAllDistricts), paste(allStates, as.character(district), sep = " "), ""))
partisanEnglish <- commissionEnglish %>%
  filter(commission == "Partisan") %>%
  mutate(outliers = ifelse(check_outlier(englishAllDistricts), paste(allStates, as.character(district), sep = " "), ""))
outlierEnglish <- rbind(independentEnglish, partisanEnglish)

library(ggrepel)
ggplot(data = outlierEnglish, aes(x = commission, y = englishAllDistricts, color = commission)) + 
     geom_boxplot() + ggtitle("Absolute Difference of Percentage Speaking English at Home") +
     theme(plot.title = element_text(hjust = 0.5)) + geom_text_repel(aes(label=outliers))

plot5 <- do.call("rbind", list(whitedf,blackdf, hispanicdf, tworacedf))
plot6 <- do.call("rbind", list(unempdf, elderlydf, tradedf, agdf))
plot7 <- do.call("rbind", list(enrolleddf, somecollegedf, vetdf))
plot8 <- do.call("rbind", list(youngdf, marrieddf, englishdf))

ggplot(plot5, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1)) +
  geom_hline(yintercept = 0) + ggtitle("Race Differences by State") + ylab("Absolute Difference") + 
  xlab("") + theme(plot.title = element_text(hjust = 0.5))

ggplot(plot6, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1)) +
  geom_hline(yintercept = 0) + ggtitle("Race Differences by State") + ylab("Absolute Difference") + 
  xlab("") + theme(plot.title = element_text(hjust = 0.5))

ggplot(plot7, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1)) +
  geom_hline(yintercept = 0) + ggtitle("Educational Congressional District Neighbor Differences by State") + ylab("Absolute Difference") + 
  xlab("") + theme(plot.title = element_text(hjust = 0.5))

ggplot(plot8, aes(x = label, y = means, color = states)) +
  geom_point(size = 4, shape = 18, position = position_dodge(width=1)) +
  geom_errorbar(aes(ymax = upperconf, ymin = lowerconf), position = position_dodge(width=1)) +
  geom_hline(yintercept = 0) + ggtitle("Social Statistic Congressional District Neighbor Differences by State") + ylab("Absolute Difference") + 
  xlab("") + theme(plot.title = element_text(hjust = 0.5))

whiteAllDistricts <- c(whitedifferencesCA, whitedifferencesWA, whitedifferencesTX, whitedifferencesPA)
blackAllDistricts <- c(blackdifferencesCA, blackdifferencesWA, blackdifferencesTX, blackdifferencesPA)
hispanicAllDistricts <- c(hispanicdifferencesCA, hispanicdifferencesWA, hispanicdifferencesTX, hispanicdifferencesPA)
tworaceAllDistricts <- c(tworacedifferencesCA, tworacedifferencesWA, tworacedifferencesTX, tworacedifferencesPA)
labels <- c(rep("Percentage White Difference", 117), rep("Percentage Black Difference", 117), rep("Percentage Hispanic Difference", 117),
            rep("Percentage Biracial Difference",117))
raceData <- c(whiteAllDistricts, blackAllDistricts, hispanicAllDistricts, tworaceAllDistricts)
allStates <- rep(c(rep("California",53), rep("Washington",10), rep("Texas",36), rep("Pennsylvania", 18)),4)
allCommission <- rep(c(rep("Independent", 63), rep("Partisan", 54)),4)
district <- rep(c(cali2012$DistrictNum, wash2012$DistrictNum, tx2012$DistrictNum, penn2012$DistrictNum),4)
raceBoxDF <- data.frame(raceData, allStates, allCommission, district, labels)

unempAllDistricts <- c(unempdifferencesCA, unempdifferencesWA, unempdifferencesTX, unempdifferencesPA)
elderlyAllDistricts <- c(elderlydifferencesCA, elderlydifferencesWA, elderlydifferencesTX, elderlydifferencesPA)
tradeAllDistricts <- c(tradedifferencesCA, tradedifferencesWA, tradedifferencesTX, tradedifferencesPA)
agAllDistricts <- c(agdifferencesCA, agdifferencesWA, agdifferencesTX, agdifferencesPA)
enrolledAllDistricts <- c(enrolleddifferencesCA, enrolleddifferencesWA, enrolleddifferencesTX, enrolleddifferencesPA)
somecollegeAllDistricts <- c(somecollegedifferencesCA, somecollegedifferencesWA, somecollegedifferencesTX, somecollegedifferencesPA)
vetAllDistricts <- c(vetdifferencesCA, vetdifferencesWA, vetdifferencesTX, vetdifferencesPA)
youngAllDistricts <- c(youngdifferencesCA, youngdifferencesWA, youngdifferencesTX, youngdifferencesPA)
marriedAllDistricts <- c(marrieddifferencesCA, marrieddifferencesWA, marrieddifferencesTX, marrieddifferencesPA)
foreignAllDistricts <- c(foreigndifferencesCA, foreigndifferencesWA, foreigndifferencesTX, foreigndifferencesPA)
incomeAllDistricts <- c(incomedifferencesCA, incomedifferencesWA, incomedifferencesTX, incomedifferencesPA)
allData <- c(raceData, unempAllDistricts, elderlyAllDistricts, tradeAllDistricts, agAllDistricts, enrolledAllDistricts, 
             somecollegeAllDistricts, vetAllDistricts, youngAllDistricts, marriedAllDistricts, englishAllDistricts, 
             foreignAllDistricts, incomeAllDistricts)

labels <- c(rep("Percentage White Difference", 117), 
            rep("Percentage Black Difference", 117), 
            rep("Percentage Hispanic Difference", 117),
            rep("Percentage Biracial Difference",117), 
            rep("Unemployment Rate Difference",117), 
            rep("Percent Poor Elderly Difference", 117),
            rep("Percent Employed in Wholesale Trade Difference", 117), 
            rep("Percent Employed in Agriculture Difference", 117), 
            rep("Percent Enrolled in College Difference", 117), 
            rep("Percent College Dropout Difference", 117), 
            rep("Veteran Population Difference", 117), 
            rep("Percent Age 20-34 Difference",117), 
            rep("Married Female Population Difference", 117),
            rep("Percent Speak English at Home Difference",117),
            rep("Foreign Born Difference",117),
            rep("Per Capita Income Difference", 117))
allStates <- rep(c(rep("California",53), rep("Washington",10), rep("Texas",36), rep("Pennsylvania", 18)),16)
allCommission <- rep(c(rep("Independent", 63), rep("Partisan", 54)),16)
district <- rep(c(cali2012$DistrictNum, wash2012$DistrictNum, tx2012$DistrictNum, penn2012$DistrictNum),16)
allBoxDF <- data.frame(allData, allStates, allCommission, district, labels)

is_outlier <- function(x) {
  return(x < quantile(x, 0.25) - 1.5 * IQR(x) | x > quantile(x, 0.75) + 1.5 * IQR(x))
}
raceBoxDF %>%
  group_by(labels, allCommission) %>%
  mutate(outlier = ifelse(is_outlier(raceData), paste(allStates, as.character(district), sep = " "), "")) %>%
  ggplot(., aes(x = labels, y = raceData, color = allCommission)) +
    geom_boxplot() + geom_text(aes(label = outlier))

ggplot(raceBoxDF, aes(x = labels, y = raceData, color = allCommission)) +
  geom_boxplot() + ggtitle("Race Differences by State") + ylab("Absolute Difference") + 
  xlab("") + theme(plot.title = element_text(hjust = 0.5))

allOutlier <- allBoxDF %>%
  group_by(labels, allCommission) %>%
  mutate(outlier = ifelse(is_outlier(allData), paste(allStates, as.character(district), sep = " "), ""))

ggplot(allOutlier, aes(x = labels, y = allData)) + geom_boxplot(aes(fill = allCommission)) + 
  ggtitle("Demographic Congressional District Differences by District Drawing Method") + ylab("Absolute Difference") + 
  xlab("") + theme(plot.title = element_text(hjust = 0.5)) + #geom_text_repel(aes(label = outlier)) +
  facet_wrap(~ labels, scales = "free")

ggplot(allOutlier, aes(x = labels, y = allData)) + geom_boxplot(aes(fill = allCommission)) + 
  ggtitle("Demographic Congressional District Differences by District Drawing Method") + ylab("Absolute Difference") + 
  xlab("") + theme(plot.title = element_text(hjust = 0.5)) +
  facet_wrap(~ labels, scales = "free")

onlyOutliers <- filter(allOutlier, outlier != "")
ggplot(onlyOutliers, aes(allStates)) + geom_bar()

outlierDistrictCount <- onlyOutliers %>%
  group_by(outlier) %>%
  summarise(n = n()) %>%
  filter(n > 1) %>%
  mutate(state= gsub('[[:digit:]]+', '', outlier))
ggplot(outlierDistrictCount, aes(outlier)) + 
  geom_bar(stat = 'identity', aes(y = n, fill=state)) +
  ggtitle("Number of Variables District is an Outlier On") +
  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5), plot.title = element_text(hjust = 0.5)) +
  xlab("District") + ylab("Number of Outliers")

NA

totalOutlierCount <- allBoxDF %>%
  group_by(labels, allStates) %>%
  mutate(outlier = ifelse(is_outlier(allData), paste(allStates, as.character(district), sep = " "), "")) %>%
  ungroup() %>%
  filter(outlier != "") %>%
  group_by(outlier) %>%
  summarise(n = n()) %>%
  filter(n > 3) %>%
  mutate(state= gsub('[[:digit:]]+', '', outlier))
ggplot(totalOutlierCount, aes(outlier)) + 
  geom_bar(stat = 'identity', aes(y = n, fill=state)) +
  ggtitle("Number of Variables District is an Outlier On (> 3)") +
  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5), plot.title = element_text(hjust = 0.5)) +
  xlab("District") + ylab("Number of Outliers")

stateOutlierProp <- allBoxDF %>%
  group_by(labels, allCommission) %>%
  mutate(outlier = ifelse(is_outlier(allData), paste(allStates, as.character(district), sep = " "), "")) %>%
  summarise(percent_outlier = ((length(unique(outlier))-1)/n()))
ggplot(stateOutlierProp, aes(allCommission)) + 
  geom_bar(stat = 'identity', aes(y = percent_outlier, fill=allCommission)) +
  ggtitle("Proportion of Districts are Outliers by District Drawing Commission") +
  theme(axis.text.x=element_text(angle=90,hjust=1,vjust=0.5), plot.title = element_text(hjust = 0.5)) +
  xlab("District") + ylab("Number of Outliers") + facet_wrap(~ labels, scales = "free")

Error: `predicate` must be a closure or function pointer

pennOutlierCount <- allOutlier %>%
  filter(allStates == "Pennsylvania") %>%
  group_by(district) %>%
  summarise(times_outlier = sum(ifelse(outlier != "",1,0))) 

outlier_pa <- tibble(DISTRICT=as.character(1:18),
                   Outliers = pennOutlierCount$times_outlier)
cd114_pa <- cd114_pa %>% left_join(outlier_pa, by="DISTRICT")
cd114_pa

Simple feature collection with 18 features and 2 fields
geometry type:  MULTIPOLYGON
dimension:      XY
bbox:           xmin: -80.51949 ymin: 39.7198 xmax: -74.68952 ymax: 42.26986
epsg (SRID):    4269
proj4string:    +proj=longlat +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +no_defs
First 10 features:
   DISTRICT Outliers                       geometry
1         1        5 MULTIPOLYGON (((-75.4183 39...
2         2        6 MULTIPOLYGON (((-75.34917 4...
3         3        0 MULTIPOLYGON (((-79.76195 4...
4         4        0 MULTIPOLYGON (((-77.47109 3...
5         5        2 MULTIPOLYGON (((-80.26614 4...
6         6        0 MULTIPOLYGON (((-76.47761 4...
7         7        0 MULTIPOLYGON (((-76.18406 4...
8         8        0 MULTIPOLYGON (((-75.55788 4...
9         9        0 MULTIPOLYGON (((-80.11704 3...
10       10        0 MULTIPOLYGON (((-77.91393 4...

devtools::install_github("tidyverse/ggplot2")
require(ggplot2)

dm <- get_map(location="Washington", zoom=6) 

Source : https://maps.googleapis.com/maps/api/staticmap?center=Washington&zoom=6&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA
Source : https://maps.googleapis.com/maps/api/geocode/json?address=Washington&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA
geocode failed with status REQUEST_DENIED, location = "Washington"Error in data.frame(ll.lat = ll[1], ll.lon = ll[2], ur.lat = ur[1], ur.lon = ur[2]) : 
  arguments imply differing number of rows: 0, 1

map<-get_map(location = c(-77.1945,41.2033), zoom=7)

Source : https://maps.googleapis.com/maps/api/staticmap?center=41.2033,-77.1945&zoom=7&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

ggmap(map) + 
  geom_sf(data=cd114_pa,aes(fill=Outliers),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "green", high = "red", limits=c(0,7)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

washOutlierCount <- allOutlier %>%
  filter(allStates == "Washington") %>%
  group_by(district) %>%
  summarise(times_outlier = sum(ifelse(outlier != "",1,0)))
cd114_wa <- cd114 %>% 
            filter(STATENAME=="Washington") %>%
            mutate(DISTRICT = as.character(DISTRICT)) %>%
            select(DISTRICT)
outlier_wa <- tibble(DISTRICT=as.character(1:10),
                   Outliers = washOutlierCount$times_outlier)
cd114_wa <- cd114_wa %>% left_join(outlier_wa, by="DISTRICT")

map<-get_map(location = c(-120.7401,47.7511), zoom=6)

Source : https://maps.googleapis.com/maps/api/staticmap?center=47.7511,-120.7401&zoom=6&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

ggmap(map) + 
  geom_sf(data=cd114_wa,aes(fill=Outliers),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "green", high = "red", limits=c(0,7)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

txOutlierCount <- allOutlier %>%
  filter(allStates == "Texas") %>%
  group_by(district) %>%
  summarise(times_outlier = sum(ifelse(outlier != "",1,0)))
cd114_tx <- cd114 %>% 
            filter(STATENAME=="Texas") %>%
            mutate(DISTRICT = as.character(DISTRICT)) %>%
            select(DISTRICT)
outlier_tx <- tibble(DISTRICT=as.character(1:36),
                   Outliers = txOutlierCount$times_outlier)
cd114_tx <- cd114_tx %>% left_join(outlier_tx, by="DISTRICT")

map<-get_map(location = c(-99.9018,31.9689), zoom=6)

Source : https://maps.googleapis.com/maps/api/staticmap?center=31.9689,-99.9018&zoom=6&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

ggmap(map) + 
  geom_sf(data=cd114_tx,aes(fill=Outliers),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "green", high = "red", limits=c(0,7)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

caOutlierCount <- allOutlier %>%
  filter(allStates == "California") %>%
  group_by(district) %>%
  summarise(times_outlier = sum(ifelse(outlier != "",1,0)))
cd114_ca <- cd114 %>% 
            filter(STATENAME=="California") %>%
            mutate(DISTRICT = as.character(DISTRICT)) %>%
            select(DISTRICT)
outlier_ca <- tibble(DISTRICT=as.character(1:53),
                   Outliers = caOutlierCount$times_outlier)
cd114_ca <- cd114_ca %>% left_join(outlier_ca, by="DISTRICT")

map<-get_map(location = c(-119.4179,36.7783), zoom=6)

Source : https://maps.googleapis.com/maps/api/staticmap?center=36.7783,-119.4179&zoom=6&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

ggmap(map) + 
  geom_sf(data=cd114_ca,aes(fill=Outliers),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "green", high = "red", limits=c(0,7)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

map_pa<-get_map(location = c(-77.1945,41.2033), zoom=7)

Source : https://maps.googleapis.com/maps/api/staticmap?center=41.2033,-77.1945&zoom=7&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

map_wa<-get_map(location = c(-120.7401,47.7511), zoom=6)

Source : https://maps.googleapis.com/maps/api/staticmap?center=47.7511,-120.7401&zoom=6&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

map_tx<-get_map(location = c(-99.9018,31.9689), zoom=6)

Source : https://maps.googleapis.com/maps/api/staticmap?center=31.9689,-99.9018&zoom=6&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

map_ca<-get_map(location = c(-119.4179,36.7783), zoom=6)

Source : https://maps.googleapis.com/maps/api/staticmap?center=36.7783,-119.4179&zoom=6&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

voteshare_pa <- tibble(DISTRICT=as.character(1:18),
                   `Average D Voteshare` = filter(avg_votes, State == "Pennsylvania")$Average)
cd114_vote_pa <- cd114_pa %>% left_join(voteshare_pa, by="DISTRICT")

ggmap(map_pa) + 
  geom_sf(data=cd114_vote_pa,aes(fill=Voteshare),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "red", high = "blue", limits=c(0,1)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

voteshare_wa <- tibble(DISTRICT=as.character(1:10),
                   `Average D Voteshare` = filter(avg_votes, State == "Washington")$Average)
cd114_vote_wa <- cd114_wa %>% left_join(voteshare_wa, by="DISTRICT")

ggmap(map_wa) + 
  geom_sf(data=cd114_vote_wa,aes(fill=Voteshare),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "red", high = "blue", limits=c(0,1)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

voteshare_ca <- tibble(DISTRICT=as.character(1:53),
                   `Average D Voteshare` = filter(avg_votes, State == "California")$Average)
cd114_vote_ca <- cd114_ca %>% left_join(voteshare_ca, by="DISTRICT")

ggmap(map_ca) + 
  geom_sf(data=cd114_vote_ca,aes(fill=Voteshare),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "red", high = "blue", limits=c(0,1)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

voteshare_tx <- tibble(DISTRICT=as.character(1:36),
                   `Average D Voteshare` = filter(avg_votes, State == "Texas")$Average)
cd114_vote_tx <- cd114_tx %>% left_join(voteshare_tx, by="DISTRICT")

ggmap(map_tx) + 
  geom_sf(data=cd114_vote_tx,aes(fill=`Average D Voteshare`),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "red", high = "blue", limits=c(0,1)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

map_philly <- get_map(c(-75.1652,39.9526), zoom=9)

Source : https://maps.googleapis.com/maps/api/staticmap?center=39.9526,-75.1652&zoom=9&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

ggmap(map_philly) + 
  geom_sf(data=cd114_pa,aes(fill=Outliers),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "green", high = "red", limits=c(0,7)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

map_la <- get_map(c(-118.2437,34.0522), zoom=8)

Source : https://maps.googleapis.com/maps/api/staticmap?center=34.0522,-118.2437&zoom=8&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

ggmap(map_la) +
  geom_sf(data=cd114_ca,aes(fill=Outliers),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "green", high = "red", limits=c(0,7)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

map_dallas <- get_map(c(-96.7970,32.7767), zoom=7)

Source : https://maps.googleapis.com/maps/api/staticmap?center=32.7767,-96.797&zoom=7&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

ggmap(map_dallas) + 
  geom_sf(data=cd114_tx,aes(fill=Outliers),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "green", high = "red", limits=c(0,7)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.

map_hou <- get_map(c(-95.3698,29.7604), zoom=8)

Source : https://maps.googleapis.com/maps/api/staticmap?center=29.7604,-95.3698&zoom=8&size=640x640&scale=2&maptype=terrain&language=en-EN&key=AIzaSyAcFP4aqbKzzeaBtK9xdyUsX3Lwam9zDAA

ggmap(map_hou) + 
  geom_sf(data=cd114_tx,aes(fill=Outliers),inherit.aes=FALSE,alpha=0.9) + 
  scale_fill_gradient(low = "green", high = "red", limits=c(0,7)) +
  theme(axis.title.x=element_blank(),
        axis.text.x=element_blank(),
        axis.ticks.x=element_blank(),
        axis.title.y=element_blank(),
        axis.text.y=element_blank(),
        axis.ticks.y=element_blank())

Coordinate system already present. Adding new coordinate system, which will replace the existing one.